Online damage identification method for cantilever beam based on single-point vibration response

Abstract

The application provides a cantilever beam online damage identification method based on single-point vibration response, comprising the following steps: acquiring a time-domain displacement response signal at a predetermined damage-sensitive position of a cantilever beam; determining a displacement response power spectral density at the damage-sensitive position based on the time-domain displacement response signal; extracting a characteristic index based on the displacement response power spectral density, and inputting the characteristic index into a pre-trained damage identification model to obtain a damage position of the cantilever beam. The application can quickly and accurately locate the damage of the cantilever beam structure based on the single-point response signal, and improves the accuracy and efficiency of damage identification.

Description

Technical Field

[0001] This invention relates to the field of engineering structural health monitoring technology, and in particular to an online damage identification method for cantilever beams based on single-point vibration response. Background Technology

[0002] Cantilever beams are one of the most common structures in engineering, with wide applications in civil engineering, aviation, and aerospace. During long-term service, cantilever beam structures are subjected to various complex loads such as vibration, impact, and aerodynamics. Over time, cracks and various damages can easily develop, reducing the structural safety performance. Timely detection, assessment, and repair of damage are of great significance for ensuring structural safety and reducing major losses.

[0003] When a cantilever beam structure suffers damage, its dynamic characteristic parameters change, allowing for the monitoring of the structure's health status. Traditional damage identification methods based on structural dynamic characteristic parameters often use natural frequencies, mode shapes, modal curvature, and modal strain energy as damage indicators. However, traditional methods have certain limitations: first, natural frequencies, as global variables of the structure, are insensitive to local damage, resulting in low identification accuracy; second, local structural variables such as modal curvature require measurement points at multiple locations on the structure, making measurement complex and inaccessible at certain locations in complex structures; and third, damage indicators such as modal strain energy require complex post-processing, hindering timely and rapid real-time online damage monitoring. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide an online damage identification method for cantilever beams based on single-point vibration response, which can quickly and accurately locate damage in cantilever beam structures online based on single-point response signals, thereby improving the accuracy and efficiency of damage identification.

[0005] To achieve the above objectives, the technical solutions adopted in the embodiments of the present invention are as follows:

[0006] In a first aspect, embodiments of the present invention provide an online damage identification method for cantilever beams based on single-point vibration response, comprising: acquiring time-domain displacement response signals at predetermined damage-sensitive locations of the cantilever beam; determining the displacement response power spectral density at the damage-sensitive locations based on the time-domain displacement response signals; extracting feature indicators based on the displacement response power spectral density, and inputting the feature indicators into a pre-trained damage identification model to obtain the damage location of the cantilever beam.

[0007] In one implementation, determining the displacement response power spectral density at the damage-sensitive location based on the time-domain displacement response signal includes: converting the time-domain displacement response signal into a frequency-domain displacement response signal using a Fourier transform; and calculating the displacement response power spectral density at the damage-sensitive location based on the frequency-domain displacement response signal.

[0008] In one implementation, the process of determining the damage-sensitive location includes: establishing a finite element model of the undamaged cantilever beam and modifying the finite element model to obtain a target finite element model; dividing the undamaged cantilever beam into multiple damage regions along the longitudinal direction of the cantilever beam structure based on different damage conditions; wherein each damage region corresponds to a damage condition; establishing damage finite element models under different damage conditions based on the target finite element model and the damage regions; performing modal analysis on the damage finite element models under different damage conditions to obtain multiple sets of structural displacement mode shape data under different damage conditions; converting the structural displacement mode shape data into modal curvature, and performing sensitivity analysis based on the modal curvature to obtain the damage-sensitive location of the cantilever beam.

[0009] In one implementation, structural displacement mode shape data is converted into modal curvature, and sensitivity analysis is performed based on the modal curvature to obtain the damage-sensitive location of the cantilever beam. This includes: converting structural displacement mode shape data into modal curvature using a second-order difference algorithm; calculating the variance of the modal curvature of the same damage region under different damage conditions; and determining the damage region with the largest variance as the damage-sensitive location of the cantilever beam.

[0010] In one implementation, the training process of the damage identification model includes: performing random vibration simulation analysis on the finite element model of damage under different damage conditions; extracting the displacement response power spectral density of the damage-sensitive location under each damage condition, and extracting feature indicators based on the displacement response power spectral density; using the feature indicators under each damage condition as input samples, and the damage area corresponding to each damage condition as output samples; and training the deep neural network model based on the input samples and output samples to obtain the trained damage identification model.

[0011] In one implementation, training a deep neural network model based on input samples and output samples includes: training the deep neural network model using cross-validation based on the input samples and output samples.

[0012] Secondly, embodiments of the present invention provide an online damage identification device for cantilever beams based on single-point vibration response, comprising: a signal acquisition module for acquiring time-domain displacement response signals at predetermined damage-sensitive locations of the cantilever beam; a power spectral density determination module for determining the displacement response power spectral density at the damage-sensitive locations based on the time-domain displacement response signals; and a damage identification module for extracting feature indicators based on the displacement response power spectral density and inputting the feature indicators into a pre-trained damage identification model to obtain the damage location of the cantilever beam.

[0013] In one embodiment, the power spectral density determination module is further configured to: convert the time-domain displacement response signal into a frequency-domain displacement response signal using Fourier transform; and calculate the displacement response power spectral density at the damage-sensitive location based on the frequency-domain displacement response signal.

[0014] Thirdly, embodiments of the present invention provide an electronic device including a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the steps of any of the methods provided in the first aspect above.

[0015] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the method provided in any of the first aspects above.

[0016] The embodiments of the present invention bring the following beneficial effects:

[0017] The online damage identification method for cantilever beams based on single-point vibration response provided in this invention first acquires the time-domain displacement response signal at a predetermined damage-sensitive location of the cantilever beam; then, it determines the displacement response power spectral density at the damage-sensitive location based on the time-domain displacement response signal; finally, it extracts feature indicators based on the displacement response power spectral density and inputs these feature indicators into a pre-trained damage identification model to obtain the damage location of the cantilever beam. This method can identify cantilever beam damage by combining the response signal at a single point (the damage-sensitive location of the cantilever beam) with a pre-trained damage identification model, thereby quickly and accurately locating the damage location of the cantilever beam, improving the accuracy and efficiency of damage identification, and ensuring the safety of the cantilever beam structure.

[0018] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0020] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0021] Figure 1 A flowchart illustrating an online damage identification method for cantilever beams based on single-point vibration response, provided in an embodiment of the present invention;

[0022] Figure 2 This is a schematic diagram of an intelligent online monitoring system for cantilever beam damage provided in an embodiment of the present invention;

[0023] Figure 3 This is a schematic diagram illustrating the damage region division of a cantilever beam structure, provided as an embodiment of the present invention.

[0024] Figure 4 A flowchart of another online damage identification method for cantilever beams based on single-point vibration response provided in an embodiment of the present invention;

[0025] Figure 5 A schematic diagram of the structure of an online damage identification device for cantilever beams based on single-point vibration response provided in an embodiment of the present invention;

[0026] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] Currently, traditional damage identification methods based on structural dynamic characteristic parameters often use natural frequencies, mode shapes, modal curvature, and modal strain energy as damage indicators. However, traditional methods have certain limitations: First, natural frequencies, as global variables of the structure, are not sensitive to local damage, resulting in low identification accuracy; second, local structural variables such as modal curvature require measurement points to be placed at multiple locations on the structure, making measurement complex and inaccessible at certain locations in complex structures; third, damage indicators such as modal strain energy require relatively complex post-processing, making timely and rapid real-time online damage monitoring impossible.

[0029] Based on this, the present invention provides an online damage identification method for cantilever beams based on single-point vibration response, which can quickly and accurately locate the damage of the cantilever beam structure online based on the single-point response signal, thereby improving the accuracy and efficiency of damage identification.

[0030] To facilitate understanding of this embodiment, a detailed description of the online damage identification method for cantilever beams based on single-point vibration response disclosed in this invention will be provided first. This method can be executed by electronic devices, such as smartphones, computers, and tablets. See also... Figure 1 The flowchart shown illustrates an online damage identification method for cantilever beams based on single-point vibration response, indicating that the method mainly includes the following steps S101 to S103:

[0031] Step S101: Obtain the time-domain displacement response signal at the predetermined damage-sensitive location of the cantilever beam.

[0032] In one implementation, see Figure 2 The diagram illustrates an intelligent online monitoring system for cantilever beam damage. Under working conditions, the cantilever beam is subjected to a random load spectrum at its root. A laser sensor is installed directly above the damage-sensitive location X of the cantilever beam to record the time-domain displacement response signal. In practical implementation, for an actual cantilever beam structure, the same load power spectrum as in the simulation can be applied to the root of the cantilever beam, and the time-domain displacement response signal at the damage-sensitive location can be measured online in real time using a laser sensor.

[0033] Step S102: Determine the displacement response power spectral density at the damage-sensitive location based on the time-domain displacement response signal.

[0034] In one implementation, the time-domain displacement response signal at the acquired damage-sensitive location can be converted into a frequency-domain displacement response signal, and then the displacement response power spectral density at the damage-sensitive location can be calculated.

[0035] Step S103: Extract feature indicators based on displacement response power spectral density, and input the feature indicators into a pre-trained damage identification model to obtain the damage location of the cantilever beam.

[0036] In one embodiment, feature indicators can be extracted based on the displacement response power spectral density, including: the first four spectral moments, peak values, and their corresponding frequencies; the extracted feature indicators are then input into a pre-trained damage identification model to obtain the damage location of the cantilever beam. In this embodiment, the damage identification model can be trained based on a deep neural network model.

[0037] The online damage identification method for cantilever beams based on single-point vibration response provided in this invention can identify cantilever beam damage by combining the response signal of a single point (the damage-sensitive location of the cantilever beam) with a pre-trained damage identification model, thereby quickly and accurately locating the damage position of the cantilever beam, improving the accuracy and efficiency of damage identification, and ensuring the safety of the cantilever beam structure. Furthermore, this method can perform damage identification in real time after obtaining the single-point response signal during the experiment, which is more timely than existing methods that require damage identification after the experiment is completed.

[0038] In one implementation, for the aforementioned step S102, i.e., when determining the displacement response power spectral density at the damage-sensitive location based on the time-domain displacement response signal, it can be implemented in ways including but not limited to the following: First, the time-domain displacement response signal is converted into a frequency-domain displacement response signal using Fourier transform; then, the displacement response power spectral density at the damage-sensitive location is calculated based on the frequency-domain displacement response signal.

[0039] In practice, the obtained time-domain displacement response signal can be converted into a frequency-domain displacement response signal by using a fast Fourier transform. Then, the square of the frequency-domain displacement response signal can be calculated and normalized to obtain the displacement response power spectral density at the damage-sensitive location.

[0040] Specifically, the frequency domain displacement response signal and displacement response power spectral density can be calculated using the following formulas:

[0041]

[0042]

[0043] Among them, u N (n) represents a set (N) discrete time-domain displacement response signals, U N (ω) represents the frequency domain displacement response signal after Fourier transform. This represents the corresponding displacement response power spectral density.

[0044] This invention also provides a method for determining damage-sensitive locations based on modal analysis, which can determine damage-sensitive locations according to the modal curvature of a cantilever beam, specifically including the following steps 1 to 5:

[0045] Step 1: Establish a finite element model of the non-destructive cantilever beam, and modify the finite element model to obtain the target finite element model.

[0046] In practical implementation, a finite element model of the non-destructive cantilever beam is first established. Then, using the experimental modal frequencies as the target and boundary parameters and material parameters as variables, a particle swarm optimization algorithm is used to modify the finite element model of the non-destructive cantilever beam, resulting in a high-precision finite element model that can replace the actual structure, i.e., the target finite element model. Specifically, during the modification, the fixed boundary conditions at the root of the cantilever beam can be equivalent to spring constraints. Using the first three experimental modal frequencies as the target and spring stiffness constraints and material parameters as optimization variables, the finite element model is modified to make it closer to the actual cantilever beam structure, thus obtaining a high-precision finite element model.

[0047] Step 2: Divide the undamaged cantilever beam into multiple damage regions along the longitudinal direction of the cantilever beam structure based on different damage conditions; each damage region corresponds to a damage condition.

[0048] In practical implementation, the longitudinal direction of the cantilever beam structure can be divided into several damage areas according to the required accuracy of structural damage identification (which can be determined according to the actual situation; the higher the accuracy, the smaller the damage area). Different damage areas represent different damage conditions, and the state without damage is also recorded as a condition. Based on this, the total number of damage conditions can be determined as N in the embodiments of the present invention.

[0049] For details, see Figure 3 The diagram illustrates the damage zone division of a cantilever beam structure. B0 represents the root of the cantilever beam, i.e., the fixed end, and 1, 2, 3, 4, 5, 6, and 7 are the seven divided damage zones. Each damage zone corresponds to a different damage condition, and the damage can be represented by stiffness reduction.

[0050] Step 3: Establish damage finite element models under different damage conditions based on the target finite element model and the damage region.

[0051] In practical implementation, corresponding finite element models can be established according to different damage conditions. That is, for a certain damage condition, the corresponding damage area is first determined, and then a damage finite element model under that damage condition is established based on the undamaged target finite element model and the corresponding damage area. For example: assuming the damage area corresponding to the second damage condition is... Figure 3 For region 2 shown, a finite element model is established for the second damage condition, where region 2 is damaged and other regions are undamaged. Other damage conditions are similar and will not be listed here.

[0052] Step 4: Perform modal analysis on the finite element model of the damage under different damage conditions to obtain multiple sets of structural displacement mode shape data under different damage conditions.

[0053] In practice, modal analysis is performed on the finite element model of the damage under each damage condition, and the displacement mode shape data at the longitudinal central axis of symmetry of the cantilever beam are extracted to obtain N sets of structural displacement mode shape data under different damage conditions.

[0054] Step 5: Convert the structural displacement mode shape data into modal curvature, and perform sensitivity analysis based on the modal curvature to obtain the damage-sensitive locations of the cantilever beam.

[0055] In practical implementation, the structural displacement mode shape data is converted into modal curvature. Sensitivity analysis is then performed using the modal curvature data to determine the most damage-sensitive location in the cantilever beam structure, denoted as the damage-sensitive location X. Specifically, the damage-sensitive location of the cantilever beam can be determined using methods including but not limited to the following: First, the structural displacement mode shape data is converted into modal curvature using a second-order difference algorithm; then, the variance of the modal curvature of the same damage region under different damage conditions is calculated; finally, the damage region with the largest variance value is determined as the damage-sensitive location of the cantilever beam.

[0056] Specifically, the modal curvature can be calculated using the following formula:

[0057]

[0058] in, This represents the modal curvature at the damage-sensitive location X. The displacement mode is represented at the damage-sensitive location X, and h represents the spacing.

[0059] After determining the damage-sensitive locations of the cantilever beam, this invention further provides a training process for a damage identification model. A deep neural network model is established based on random vibration analysis, and the corresponding signals from the damage-sensitive locations of the cantilever beam are used as sample data to train the damage identification model. Specifically, this includes the following steps (1) to (4):

[0060] Step (1): Perform random vibration simulation analysis on the finite element model of damage under different damage conditions.

[0061] In practical implementation, the load power spectrum of the cantilever beam under working conditions can be applied to the root of the cantilever beam, and random vibration simulation analysis can be performed on the finite element model of the cantilever beam damage under N different damage conditions.

[0062] Step (2): Extract the displacement response power spectral density of the damage-sensitive location under each damage condition, and extract feature indicators based on the displacement response power spectral density.

[0063] In practical implementation, during the simulation process, the displacement response power spectral density at the damage-sensitive location X under each damage condition can be extracted. For N damage conditions, N sets of displacement response power spectral densities can be obtained, and then feature indicators can be extracted from the displacement response power spectral density.

[0064] Step (3): Use the characteristic indicators of each damage condition as input samples and the damage area corresponding to each damage condition as output samples.

[0065] Step (4): Train the deep neural network model based on the input and output samples to obtain the trained damage recognition model.

[0066] In practical implementation, the characteristic indicators under each damage condition are used as input samples, and the damage area corresponding to each damage condition is used as output samples. Based on the input and output samples, the number of layers and the number of nodes in each layer of the deep neural network are determined. A training regression deep neural network model is built and trained using cross-validation to obtain a trained damage identification model. Subsequently, online health monitoring of cantilever beam structures can be carried out based on this damage identification model.

[0067] For ease of understanding, this embodiment of the invention also provides a flowchart of another online damage identification method for cantilever beams based on single-point vibration response, see [link to flowchart]. Figure 4 As shown, this method mainly consists of three parts:

[0068] (1) Determine the damage-sensitive location based on modal analysis.

[0069] Specifically, this part mainly includes the following steps 1 to 4:

[0070] Step 1: Establish a finite element model of the non-destructive cantilever beam, and equate the fixed boundary conditions at the root to spring constraints; take the first three experimental modal frequencies as the target, and use spring stiffness constraints and material parameters as optimization variables, and use the particle swarm optimization algorithm to correct the finite element model of the non-destructive cantilever beam structure to obtain a high-precision finite element model that can replace the actual structure.

[0071] Step 2: Based on the required accuracy for structural damage identification, divide the cantilever beam structure into several damage regions along its longitudinal direction. The region division can be appropriately denser in areas prone to damage and sparser in areas less prone to damage, based on prior knowledge. Different damage regions represent different damage conditions, and the state without damage is also recorded as a condition. The total number of damage conditions is determined to be N.

[0072] Damage can be represented by a uniformly set stiffness reduction factor. When the stiffness change of the actual structure reaches this reduction factor, damage is considered to have occurred at that point.

[0073] Step 3: Establish corresponding finite element models according to different damage conditions, perform modal analysis respectively, and extract the first-order displacement mode shape along the longitudinal central symmetry axis of the cantilever beam to obtain N sets of structural displacement mode shape data under different damage conditions.

[0074] Specifically, damage finite element models are established for each damage condition, and modal analysis is performed on the damage finite element models for each damage condition. The first-order displacement mode shape data at the longitudinal central axis of symmetry of the cantilever beam are extracted to obtain N sets of structural displacement mode shape data under different damage conditions.

[0075] Step 4: Use second-order difference to convert the structural displacement mode shape data into modal curvature, and determine the most damage-sensitive location in the structure based on the modal curvature data under different damage conditions, i.e., the damage-sensitive location.

[0076] Specifically, the variance of the modal curvature at each location (i.e., each damaged area) under different damage conditions is used as an indicator. The larger the variance, the more sensitive it is to damage. The location with the largest variance is the most sensitive location to damage, denoted as the damage-sensitive location X.

[0077] (2) A deep neural network model was established based on random vibration analysis.

[0078] Specifically, this section mainly includes the following steps 5 to 7:

[0079] Step 5: Apply the load power spectrum of the cantilever beam under working conditions to the root of the cantilever beam, and perform random vibration simulation analysis on the cantilever beam under N different damage conditions. Based on the random vibration simulation results, extract the displacement response power spectral density G at the damage-sensitive location X. X (f).

[0080] Step 6: Extract feature indices (including the first four spectral moments, peak values ​​and their corresponding frequencies) from the displacement response power spectral density as input samples, and use the location of the damaged area of ​​the divided cantilever beam structure as the output.

[0081] Specifically, N sets of displacement response power spectral densities were obtained based on random vibration analysis under different damage conditions. Feature indices were extracted from these N sets of displacement response power spectral densities as input samples, and the location of the damaged area was used as the output sample. Notably, the damage location corresponding to the no-damage state was 0. The formula for calculating the i-th order spectral moment is:

[0082]

[0083] Where f represents frequency, m i Let G represent the i-th order spectral moment. X (f) represents the power spectral density of the unilateral displacement response.

[0084] Step 7: Based on the input and output samples, determine the number of layers and nodes in each layer of the deep neural network. Use cross-validation to build and train a regression-type deep neural network model. This deep neural network model can then be used for online health monitoring of cantilever beam structures.

[0085] (3) Online damage identification of cantilever beams.

[0086] Specifically, this section mainly includes the following steps 8 to 10:

[0087] Step 8: For the actual cantilever beam structure, apply the same load power spectrum as in the simulation to the root of the cantilever beam, and install the sensor at the damage-sensitive location X to measure the time-domain displacement response signal at the damage-sensitive location X in real time online.

[0088] Step 9: For the measured time-domain displacement response signal at the damage-sensitive location X, use Fast Fourier Transform to convert the time-domain displacement response signal to the frequency domain, and then calculate the displacement response power spectral density at the damage-sensitive location X.

[0089] Step 10: Extract the first four spectral moments, peak values, and their corresponding frequencies from the displacement response power spectral density as inputs, and substitute them into the trained deep neural network model to quickly and accurately identify and locate damage.

[0090] The method provided in this embodiment of the invention can accurately locate the damage position of a structure based on the measurement information of a single sensor, and has at least the following beneficial effects:

[0091] (1) For damage monitoring of cantilever beams, the location most sensitive to damage can be identified. Without the need to deploy a series of sensors, the health status of the cantilever beam structure can be conveniently determined based solely on the single-point measured response signal at the damage-sensitive location.

[0092] (2) Based on random vibration response data, feature indicators are extracted as samples, and a deep neural network model is built and trained using cross-validation method, which can accurately and quickly locate structural damage online.

[0093] (3) Once the deep neural network model in the embodiments of the present invention is established, the location of structural damage can be quickly determined based on the data measured by the laser sensor at a single point, and the health status of the cantilever beam structure can be monitored online in real time.

[0094] In addition to the aforementioned online damage identification method for cantilever beams based on single-point vibration response, this invention also provides an online damage identification device for cantilever beams based on single-point vibration response. (See attached document.) Figure 5 The diagram shows a structural schematic of an online damage identification device for cantilever beams based on single-point vibration response. It illustrates that the device mainly comprises the following parts:

[0095] The signal acquisition module 501 is used to acquire the time-domain displacement response signal at the predetermined damage-sensitive location of the cantilever beam;

[0096] The power spectral density determination module 502 is used to determine the displacement response power spectral density at the damage-sensitive location based on the time-domain displacement response signal.

[0097] The damage identification module 503 is used to extract feature indicators based on the displacement response power spectral density and input the feature indicators into a pre-trained damage identification model to obtain the damage location of the cantilever beam.

[0098] The online damage identification device for cantilever beams based on single-point vibration response provided in this embodiment of the invention can identify damage to cantilever beams by combining the response signal of a single point (the damage-sensitive location of the cantilever beam) with a pre-trained damage identification model, thereby quickly and accurately locating the damage location of the cantilever beam, improving the accuracy and efficiency of damage identification, and ensuring the safety of the cantilever beam structure.

[0099] In one embodiment, the power spectral density determination module 502 is further configured to: convert the time-domain displacement response signal into a frequency-domain displacement response signal using Fourier transform; and calculate the displacement response power spectral density at the damage-sensitive location based on the frequency-domain displacement response signal.

[0100] In one embodiment, the above-mentioned device further includes a damage-sensitive location determination module, used for: establishing a finite element model of the undamaged cantilever beam and modifying the finite element model to obtain a target finite element model; dividing the undamaged cantilever beam into multiple damage regions along the longitudinal direction of the cantilever beam structure based on different damage conditions; wherein each damage region corresponds to a damage condition; establishing damage finite element models under different damage conditions based on the target finite element model and the damage regions; performing modal analysis on the damage finite element models under different damage conditions to obtain multiple sets of structural displacement mode shape data under different damage conditions; converting the structural displacement mode shape data into modal curvature, and performing sensitivity analysis based on the modal curvature to obtain the damage-sensitive location of the cantilever beam.

[0101] In one embodiment, the aforementioned damage-sensitive location determination module is further configured to: convert structural displacement mode shape data into modal curvature using a second-order difference algorithm; calculate the variance of modal curvature of the same damage region under different damage conditions; and determine the damage region with the largest variance as the damage-sensitive location of the cantilever beam.

[0102] In one embodiment, the above-mentioned device further includes a model training module, used for: performing random vibration simulation analysis on the damage finite element model under different damage conditions; extracting the displacement response power spectral density of the damage-sensitive location under each damage condition, and extracting feature indicators based on the displacement response power spectral density; using the feature indicators under each damage condition as input samples, and the damage area corresponding to each damage condition as output samples; and training the deep neural network model based on the input samples and output samples to obtain a trained damage recognition model.

[0103] In one implementation, the model training module is further configured to: train the deep neural network model using cross-validation based on the input samples and output samples.

[0104] The device provided in this embodiment of the invention has the same implementation principle and technical effect as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0105] This invention also provides an electronic device, specifically, the electronic device includes a processor and a storage device; the storage device stores a computer program, and the computer program, when run by the processor, executes the method described in any of the above embodiments.

[0106] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. The electronic device 100 includes: a processor 60, a memory 61, a bus 62, and a communication interface 63. The processor 60, the communication interface 63, and the memory 61 are connected through the bus 62. The processor 60 is used to execute executable modules, such as computer programs, stored in the memory 61.

[0107] The memory 61 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 63 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.

[0108] Bus 62 can be an ISA bus, LXI bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 6 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0109] The memory 61 is used to store programs. After receiving an execution instruction, the processor 60 executes the program. The method executed by the device for defining the flow process disclosed in any of the foregoing embodiments of the present invention can be applied to the processor 60 or implemented by the processor 60.

[0110] Processor 60 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 60 or by instructions in software form. Processor 60 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 61. Processor 60 reads the information in memory 61 and, in conjunction with its hardware, completes the steps of the above method.

[0111] The computer program product of the readable storage medium provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the foregoing method embodiments. For specific implementation, please refer to the foregoing method embodiments, which will not be repeated here.

[0112] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0113] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for online damage identification of cantilever beams based on single-point vibration response, characterized in that, include: Obtain the time-domain displacement response signal at the predetermined damage-sensitive location of the cantilever beam; The displacement response power spectral density at the damage-sensitive location is determined based on the time-domain displacement response signal. Feature indices are extracted based on the displacement response power spectral density, and the feature indices are input into a pre-trained damage identification model to obtain the damage location of the cantilever beam. The process of determining the damage-sensitive location includes: establishing a finite element model of the undamaged cantilever beam and modifying the finite element model to obtain a target finite element model; dividing the undamaged cantilever beam into multiple damage regions along the longitudinal direction of the cantilever beam structure based on different damage conditions; wherein each damage region corresponds to a damage condition; establishing damage finite element models under different damage conditions based on the target finite element model and the damage regions; performing modal analysis on the damage finite element models under different damage conditions to obtain multiple sets of structural displacement mode shape data under different damage conditions; converting the structural displacement mode shape data into modal curvature, and performing sensitivity analysis based on the modal curvature to obtain the damage-sensitive location of the cantilever beam; The process of converting the structural displacement mode shape data into modal curvature and performing sensitivity analysis based on the modal curvature to obtain the damage-sensitive location of the cantilever beam includes: using a second-order difference algorithm to convert the structural displacement mode shape data into modal curvature; calculating the variance of the modal curvature of the same damage region under different damage conditions; and determining the damage region with the largest variance value as the damage-sensitive location of the cantilever beam.

2. The method according to claim 1, characterized in that, Determining the displacement response power spectral density at the damage-sensitive location based on the time-domain displacement response signal includes: The time-domain displacement response signal is converted into a frequency-domain displacement response signal using Fourier transform; The displacement response power spectral density at the damage-sensitive location is calculated based on the frequency domain displacement response signal.

3. The method according to claim 1, characterized in that, The training process of the damage recognition model includes: Random vibration simulation analysis was performed on the finite element model of the damage under different damage conditions. Extract the displacement response power spectral density of the damage-sensitive location under each of the aforementioned damage conditions, and extract feature indices based on the displacement response power spectral density; The characteristic indicators of each of the aforementioned damage conditions are used as input samples, and the damage area corresponding to each of the aforementioned damage conditions is used as output samples. The deep neural network model is trained based on the input samples and the output samples to obtain a trained damage recognition model.

4. The method according to claim 3, characterized in that, Training a deep neural network model based on the input samples and the output samples includes: The deep neural network model is trained using cross-validation based on the input samples and the output samples.

5. An online damage identification device for cantilever beams based on single-point vibration response, characterized in that, include: The signal acquisition module is used to acquire the time-domain displacement response signal at the predetermined damage-sensitive location of the cantilever beam; A power spectral density determination module is used to determine the displacement response power spectral density at the damage-sensitive location based on the time-domain displacement response signal. The damage identification module is used to extract feature indicators based on the displacement response power spectral density and input the feature indicators into a pre-trained damage identification model to obtain the damage location of the cantilever beam. The device further includes a damage-sensitive location determination module, used for: establishing a finite element model of the undamaged cantilever beam and modifying the finite element model to obtain a target finite element model; dividing the undamaged cantilever beam into multiple damage regions along the longitudinal direction of the cantilever beam structure based on different damage conditions; wherein each damage region corresponds to a damage condition; establishing damage finite element models under different damage conditions based on the target finite element model and the damage regions; performing modal analysis on the damage finite element models under different damage conditions to obtain multiple sets of structural displacement mode shape data under different damage conditions; converting the structural displacement mode shape data into modal curvature, and performing sensitivity analysis based on the modal curvature to obtain the damage-sensitive location of the cantilever beam; The damage-sensitive location determination module is further configured to: convert the structural displacement mode shape data into modal curvature using a second-order difference algorithm; calculate the variance of the modal curvature of the same damage region under different damage conditions; and determine the damage region with the largest variance value as the damage-sensitive location of the cantilever beam.

6. The apparatus according to claim 5, characterized in that, The power spectral density determination module is also used for: The time-domain displacement response signal is converted into a frequency-domain displacement response signal using Fourier transform; The displacement response power spectral density at the damage-sensitive location is calculated based on the frequency domain displacement response signal.

7. An electronic device, characterized in that, The method includes a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the steps of the method according to any one of claims 1 to 4.

8. A computer-readable storage medium storing a computer program thereon, characterized in that, The computer program is executed by the processor to perform the steps of the method described in any one of claims 1 to 4.

Images